Clustering and latent structure · MLD-0821
Novelty detection
Detecting new observations unlike a reference dataset assumed mostly normal.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Flag unfamiliar behavior after training on normal activity.
Original caveat
Training-set outlier detection is a related but different setup.
Context and aliases
Level: Intermediate. Stages: Training · Data · Evaluation.
Tags: ML · DL.
Aliases: None recorded.
Related concepts
Concepts linking here
Original references
- scikit-learn: clustering ↗ · Reference link
- scikit-learn: matrix decomposition ↗ · Reference link
- scikit-learn: manifold learning ↗ · Reference link
- scikit-learn: novelty and outlier detection ↗ · Reference link
- Original reference URL ↗